article · Sensors
Wearing face masks has become common across sectors such as healthcare, transport, construction, and public safety, creating a demand for reliable identity verification when facial features are partially obscured. A hybrid masked face recognition system addresses this by combining multiple computational techniques. The method incorporates a pretrained ssd-MobileNetV2 model to detect mask presence and placement, alongside landmark and oval face detection algorithms to locate vital facial features. Robust principal component analysis is deployed to isolate occluded areas from visible facial components, reducing interference caused by masks. Classification relies on a k-nearest neighbours algorithm, refined through particle swarm optimisation to fine-tune both feature selection and parameter values. Tested experimentally, the complete pipeline attained a 97 percent recognition accuracy, outperforming existing benchmarks and demonstrating high resilience against facial occlusion.
Standard facial recognition tools frequently fail when individuals wear protective coverings. By reliably identifying people without requiring them to remove masks, this technique supports security, access control, and authentication across diverse settings. This capability is particularly relevant for environments like hospitals, transport hubs, and manufacturing facilities where mask use remains standard practice for safety or hygiene.
The system offers an applied and tested approach to identity authentication and access control. Relevant users span operators in healthcare, retail, public transport, manufacturing, and security services requiring non-intrusive identification of masked personnel or visitors. Having achieved 97 percent accuracy in experimental evaluations, the core algorithm appears technically validated, though operational deployment would require integration into existing video surveillance hardware or access management software platforms.
AI-generated from the published abstract. Always read the original work before citing.
Face masks are widely used in various industries and jobs, such as healthcare, food service, construction, manufacturing, retail, hospitality, transportation, education, and public safety. Masked face recognition is essential to accurately identify and authenticate individuals wearing masks. Masked face recognition has emerged as a vital technology to address this problem and enable accurate identification and authentication in masked scenarios. In this paper, we propose a novel method that utilizes a combination of deep-learning-based mask detection, landmark and oval face detection, and robust principal component analysis (RPCA) for masked face recognition. Specifically, we use pretrained ssd-MobileNetV2 for detecting the presence and location of masks on a face and employ landmark and oval face detection to identify key facial features. The proposed method also utilizes RPCA to separate occluded and non-occluded components of an image, making it more reliable in identifying faces with masks. To optimize the performance of our proposed method, we use particle swarm optimization (PSO) to optimize both the KNN features and the number of k for KNN. Experimental results demonstrate that our proposed method outperforms existing methods in terms of accuracy and robustness to occlusion. Our proposed method achieves a recognition rate of 97%, which is significantly higher than the state-of-the-art methods. Our proposed method represents a significant improvement over existing methods for masked face recognition, providing high accuracy and robustness to occlusion.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/s23156727
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
